I think that RISC OS can be successfull on the mobile device market. I mean: We do not need to beat windows. If we would reach 1% of the market than this would be really a great success. RISC OS has the advantage that it is a full desktop OS that could run on a mobile device. So you can have a full featured text processor. I have a Windows CE Samrtphone with 500 MHz ARM processor - and I like it because it is much better than any phone I had before. But e.g. the Word version that comes with it is rather limited. On this phone it doesn't matter because you do not have a keyboard anyway. But there are mobile devices with keyboards where you would see that Pocket Word is very limited. I understand that RISC OS can not be used on a phone because it is no real time OS but I think on mobile devices we would have a chance. Look at the EeePC to see that a mobile device does not neet to run RISC OS to be successfull.
Anyway, I think that we will probably not see a native RISC OS computer after Iyonix / A9home. I hope we will but we should get used to the idea that there will be no native hardware in development. But: There are some ARM computer produced, as mobile devices as well as desktop computers (most embedded but comparable to the A9 I would say). So anyway it would be good to establish a way to port RISC OS to existing ARM devices. Sometimes you have the problem that the hardware is not dokumented but if I think about the Android platform comming up (the Google phone platform) then this might be a chance. AFAIK the platform is based on Linux which would it make possible to have a look at the drivers for the hardware.